No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
The Second Dimension of the Supreme Court
Abstract
Describing the justices of the Supreme Court as ‘liberals’ and ‘conservatives’ has become so standard — and the left-right division on the Court is considered so entrenched — that any deviation from that pattern is treated with surprise. Attentive Court watchers know that the justices are not just politicians in robes, deciding each case on a purely ideological basis. Yet the increasingly influential empirical legal studies literature assumes just that — that a left-right ideological dimension fully describes the Supreme Court. We show that there is a second, more legally-focused dimension of judicial decision-making. A continuum between legalism and pragmatism also divides the justices, in ways that cuts against ideological preferences. The second dimension is systematic and significant, occurring in multiple different legal areas, and in consistent patterns. Seen in this way, the justices and their decisions can be understood in more complex terms, not just as ideological flag bearers, but as jurists who regularly have to choose between legal methodology and outcome preferences. In two dimensions, different patterns of coalitions emerge: in the second dimension, it is the Chief Justice and Justice Sotomayor, not Justice Kennedy, who sit at the median of the Court, and decide the balance of power.
... This nicely separates the left branch of the Court from the right branch, but it does not reveal finer structure within these branches other than extremism-versus-centrism. In particular, one cannot really see, let alone precisely measure, the degree to which these two main political branches divide into smaller attitudinal branches. Including a second dimension of the Supreme Court for ideal point estimation, as in Fischman and Jacobi (2016) and Fischman (2019), provides valuable additional information about ideology that is orthogonal to politics, but it is still not clear how one could extract divisions among the justices, quantitatively or qualitatively, from this representation. What might be needed is network theory, an important part of data science that has had far less of an impact on judicial scholarship than ideal point estimation. 1 There are a variety of ways to perform ideal point estimation. ...
... The answer is that for ideal points the meaning of these extra dimensions are the same for all justices on a Court, whereas in the tree approach each edge coming off the major axis indicates a voting deviation in a non liberal-conservative direction, but these directions need not be the same for all justices. Interpreting dimensions beyond the primary political one is a very delicate matter, as can be seen in Fischman and Jacobi (2016) and Fischman (2019) where doing so is very sensitive to the legal issues in the cases under consideration. If one is less interested in a specific second dimension and instead wishes to see what role ideological directions beyond liberal-conservative play in judicial coalition formation, regardless of what these directions are, then tree methods may be more appropriate. ...
... This distinction sounds very subtle at first, but here is what it means. Perhaps (hypothetically, for the sake of discussion) White has a strong view on the pragmatism versus legalism axis found in Fischman and Jacobi (2016). If Clark has a strong view on that same axis in the same direction as White, then rather than two long external edges sprouting from the major axis we would instead see a single internal edge sprouting from the major axis then a further split into two shorter external edges (similar to the Brennan-Warren branch). ...
The popular method of ideal point estimation provides empirical legal scholars with spatial representations of the Supreme Court justices that help elucidate ideological inclinations and voting behavior. This is done primarily in one dimension, where politics dominates, though recent work details a second dimension capturing differing attitudes on the authority of various legal actors. This paper explores a new network-theoretic tree-based method for visualizing the relationships between the justices, based on their voting records, that allows scholars to study the intricate branching structure of the Court. It is shown how this tool can be used to uncover periods in the Court’s history where the balance on the bench fractured in unusual and interesting ways. Moreover, by defining several tree-based measures and charting their evolution over time, a picture emerges that throughout the past fifty years the Court became increasingly linear and bipolar, dividing along political lines.
... Unsurprisingly, there is much more to Supreme Court voting behavior than simply onedimensional political alignment. Exploring the second dimension of ideal point estimation produced from MDS has been a recent topic of interest [5][6][7][8]. Two-dimensional MDS creates an axis orthogonal to the main political axis to accommodate voting behavior that deviates from the expected liberal-conservative spectrum, but it forces the same axis on all the justices. Trees, in contrast, provide an additional flexibility by allowing divisions into an unlimited number of additional locally determined directions. ...
... Another ideal point estimation method was developed by Peress in 2009, and he said that interpreting the second dimension of their scores is a "difficult task", but that there is "some evidence to suggest that the second dimension captures 'judicial activism' ( [22], p. 286). The first in-depth analysis of a second dimension for Supreme Court ideal point estimation came in a 2016 paper by Fischman and Jacobi and a sequel paper in 2019 by Fischman [5,6]. These two papers collected and analyzed cases for which the justices voted in an unusual pattern from a one-dimensional political perspective to see if the second dimension, as generated by MDS, could help make sense of the cases-and if, in turn, these cases could help make sense of the second MDS dimension. ...
... Put another way: MDS is compatible with the coordinate linear projection R d → R d for any d < d. In particular, the x-axis in the recent investigations of two-dimensional MDS ideal point estimation for the Supreme Court has the same interpretation (and even the same values) as the single axis in the more traditional one-dimensional MDS ideal point estimation-and, with remarkable consistency, this axis seems to capture politics in the sense of a liberal-to-conservative spectrum [5], pp. 1693-1694. ...
The voting patterns of the nine justices on the United States Supreme Court continue to fascinate and perplex observers of the Court. While it is commonly understood that the division of the justices into a liberal branch and a conservative branch inevitably drives many case outcomes, there are finer, less transparent divisions within these two main branches that have proven difficult to extract empirically. This study imports methods from evolutionary biology to help illuminate the intricate and often overlooked branching structure of the justices’ voting behavior. Specifically, phylogenetic tree estimation based on voting disagreement rates is used to extend ideal point estimation to the non-Euclidean setting of hyperbolic metrics. After introducing this framework, comparing it to one- and two-dimensional multidimensional scaling, and arguing that it flexibly captures important higher-dimensional voting behavior, a handful of potential ways to apply this tool are presented. The emphasis throughout is on interpreting these judicial trees and extracting qualitative insights from them.
... But how does one then make sense of scrambled situations like the 5-to-4 vote in Ginzburg v. The dimensionality of the Supreme Court has been an active topic of exploration among scholars [3,19,5,12], and indeed some of the majority/minority divisions that seem illogical from the political spectrum perspective naturally reveal themselves on a second dimension [8]. However, a two-dimensional binary outcome voting model based on Euclidean distance preference functions still only allows for votes in which the majority is separated from the minority by a line in the plane, and while many cases do nicely exhibit this structure, many yet do not (including, for instance, the scrambled 5-4 vote mentioned above). ...
... While the exact coordinates of ideal points vary across the different estimation methods, in one dimension (d = 1) the order of the nine justices for each natural court is remarkably consistent across the different methods and is generally compatible with expert opinion of the political leanings of the justices [8, pp.1693-4]-so one can view the first dimension in all these methods as quantifying the liberal-to-conservative spectrum. 4 The second dimension for MDS ideal point estimation has been studied in [8], where a legal interpretation is proffered: the y-axis often roughly tracks with the judicial philosophies of legalism versus pragmatism. We do not require this interpretation for what follows, but the reader may find it nonetheless enriching to consider and so is encouraged to read [8]. ...
... 4 The second dimension for MDS ideal point estimation has been studied in [8], where a legal interpretation is proffered: the y-axis often roughly tracks with the judicial philosophies of legalism versus pragmatism. We do not require this interpretation for what follows, but the reader may find it nonetheless enriching to consider and so is encouraged to read [8]. The top-left plot's minority/majority division is one that occurred frequently: it clearly splits along left/right political lines, and here Kennedy sides with the conservatives (which he did more often than not). ...
We use the United States Supreme Court as an illuminative context in which to discuss three different spatial voting preference models: an instance of the widely used single-peaked preferences, and two models that are more novel in which vote outcomes have a strength in addition to a location. We introduce each model from a formal axiomatic perspective, briefly discuss practical motivation for each in terms of judicial behavior, prove mathematical relationships among the voting coalitions compatible with each model, and then study the two-dimensional setting by presenting computational tools for working with the models and by exploring these with judicial voting data from the Supreme Court.
... But how does one then make sense of scrambled situations like the 5-to-4 vote in Ginzburg v. The dimensionality of the Supreme Court has been an active topic of exploration among scholars [3,19,5,12], and indeed some of the majority/minority divisions that seem illogical from the political spectrum perspective naturally reveal themselves on a second dimension [8]. However, a two-dimensional binary outcome voting model based on Euclidean distance preference functions still only allows for votes in which the majority is separated from the minority by a line in the plane, and while many cases do nicely exhibit this structure, many yet do not (including, for instance, the scrambled 5-4 vote mentioned above). ...
... While the exact coordinates of ideal points vary across the different estimation methods, in one dimension (d = 1) the order of the nine justices for each natural court is remarkably consistent across the different methods and is generally compatible with expert opinion of the political leanings of the justices [8, pp.1693-4]-so one can view the first dimension in all these methods as quantifying the liberal-to-conservative spectrum. 4 The second dimension for MDS ideal point estimation has been studied in [8], where a legal interpretation is proffered: the y-axis often roughly tracks with the judicial philosophies of legalism versus pragmatism. We do not require this interpretation for what follows, but the reader may find it nonetheless enriching to consider and so is encouraged to read [8]. ...
... 4 The second dimension for MDS ideal point estimation has been studied in [8], where a legal interpretation is proffered: the y-axis often roughly tracks with the judicial philosophies of legalism versus pragmatism. We do not require this interpretation for what follows, but the reader may find it nonetheless enriching to consider and so is encouraged to read [8]. The top-left plot's minority/majority division is one that occurred frequently: it clearly splits along left/right political lines, and here Kennedy sides with the conservatives (which he did more often than not). ...
We use the United States Supreme Court as an illuminative context in which to discuss three different spatial voting preference models: an instance of the widely used single-peaked preferences, and two models that are more novel in which vote outcomes have a strength in addition to a location. We introduce each model from a formal axiomatic perspective, briefly discuss practical motivation for each in terms of judicial behavior, prove mathematical relationships among the voting coalitions compatible with each model, and then study the two-dimensional setting by presenting computational tools for working with the models and by exploring these with judicial voting data from the Supreme Court.
... Arguably the most in-depth analysis of the second MDS dimension of the Supreme Court is the aptly named 2016 paper by Fischman and Jacobi, which mostly focuses on the Roberts Court [18]. They point out that since MDS is based on voting agreement rates it obviates the need to determine the ideological direction of rulings, a notoriously subtle and subjective endeavor that has continually plagued quantitative Supreme Court analysis. ...
... In the mathematics literature, specifically in discrete geometry, these half-plane coalitions are called "k-sets" for k = 5 [10]. This provides another two-dimensional spatial model of Supreme Court voting-the key here is that, in contrast to the work of Fischman and Jacobi [18] where only vertical and horizontal lines are considered, lines of arbitrary slope are allowed and thereby split the justices along a combination of the traits characterized by each dimension. As with the Voronoi coalitions, we are interested in how these half-plane coalitions compare with, and help explain, the voting coalitions. ...
... What distinguishes Breyer from the majority is his vertical distance from the center of the court. One could interpret the vertical axis as in [18] to try to explain Breyer's vote, but even without a legalistic interpretation we still see a compelling story geometrically: the five majority justices have a "common ground," so to speak, which is the trapezoidal Voronoi region depicted in Figure 7 just north of the center of the court. Thus, the majority is not the five most central justices in any one-dimensional scale, nor even the five most central justices in our two-dimensional MDS space (since that would include Breyer and lose Ginsburg), but it is the five justices most closely clustered around an idealogical point just north of the center of the Court. ...
While the U.S. Supreme Court is commonly viewed as comprising a liberal bloc and a conservative bloc, with a possible swing vote or median justice between them, surprisingly many case decisions are not explained by this simple model. We introduce a pair of spatial methods for conceptualizing many 5-to-4 voting alignments that have occurred on the Court and which defy the usual liberal/conservative dichotomy. These methods, utilizing higher order Voronoi diagrams and halving lines (k-sets), are based on the geometry of the two-dimensional ideal space locations obtained from applying multidimensional scaling to voting data. We also introduce a two-dimensional metric method for determining the crucial fifth vote in each 5-to-4 ruling and for determining the median justice in any collection of terms within a natural court.
... Unguided by priors, an IRT model can misinterpret noisy data in potentially dramatic ways (Bailey 2016, 10). In particular, the assumption that the structure of disagreement is onedimensional is not always appropriate; indeed, this assumption is contested even in the United States where the IRT approach is dominant (see Fischman and Jacobi 2016). In the case of Bosnia, previous empirical research provides good reason to suspect that ethno-national politics will be the main source of disagreement on the Court (see Schwartz and Murchison 2016), but there is no extant science about the extent to which other ideological factors-including more conventional left-right politics-may compete to drive disagreement on the Court. ...
... For the same reasons noted earlier, I restrict the MDS analysis to abstract review cases. This should also help reduce the inevitable noise that results from lumping many legal issues together (see Fischman and Jacobi 2016). Moreover, these are the cases that, according to previous research (Schwartz and Murchison 2016), are most likely to divide the domestic judges along ethno-national lines and, consequently, benefit from the supposed neutrality of the foreign judges. ...
Hybrid constitutional courts are associated with deeply divided and post-conflict contexts where the impartiality of the domestic judiciary is suspect. Such courts enlist international (i.e., foreign) judges to create an ostensibly neutral counterbalance to the presumed political biases of local judges. This mixed-methods case study of the Constitutional Court of Bosnia-Herzegovina questions the value of these hybrid courts. Contrary to what might be expected, the results of multidimensional scaling indicate that Bosnia's foreign judges have not provided a reliable counterbalance to apparent ethno-national divisions on the Court. Furthermore, qualitative analysis suggests that the foreign judges have contributed to several strategic mistakes that have probably harmed the Court's tenuous authority. It is also suggested that the presence of international judges on constitutional courts may actually discourage the kind of strategic behavior that is needed to build and sustain judicial power, particularly in deeply divided and post-conflict contexts.
... In our methodology this is not considered multidimensionality, but rather taking different positions along a single bipolar dimension (liberal/conservative). Others have combined the bipolar ideological dimension with judges' legal doctrine (legalism/pragmatism) or respect for selected legal principles (Fischman, Jacobi, 2016;Bailey, Maltzman 2008Bailey, 2013). While these may be important factors contributing to judges' decisions, they do not constitute a separate ideological dimension. ...
The article presents a conceptual framework and empirical methodology of an on-going research on the role of ideology in the decisions of the Slovenian Constitution Court. The literature review demonstrates that research on judicial ideology in the courts of European countries and international courts is still rare. This can be explained by conceptual, methodological and empirical challenges posed by this type of research. The article hence advances a conceptual framework which is, contra to the mainstream theoretical approach in the field, based on a multidimensional conception of ideology that is empirically operationalised along the economic, social and authoritarian dimensions with five possible ideological positions on each dimension. By applying the newly developed methodology to a sample of Court’s decisions, it is demonstrated that this methodological approach is able to account for ideological differences between judges. This confirms that (judges’) ideology is a complex multidimensional set of values and convictions that cannot be reduced to simply equating ideology with (possible) political affiliations.
IRT models are among the most commonly used latent trait models in all of political science, particularly in the estimation of ideal points of political actors in institutions. While widely used, IRT models are often misapplied, and a key element of their estimation, the item parameters, are almost always ignored and discarded. In this paper, we look into the application of IRT models to the estimation of judicial ideology scores by Martin and Quinn (2002). Building off of a replication and extension of Martin and Quinn (2002), we demonstrate that the often-ignored item parameters are, in fact, inconsistent with the assumptions of IRTs. Then, using a post-estimation fix that is designed to ameliorate the problem, we run the model again, generating new scores. We then compare our new ideal points to the existing ideal points and discuss the implications for both ideal point modeling generally and in judicial politics specifically. We conclude by replicating a prominent study in judicial politics that demonstrates how inconsistencies in the estimation of IRT models can be consequential and bring up concerns with the implications for what this could mean for the usefulness of scores estimated via IRT models.
How would judges compose judicial panels, if they could? We focus on a procedure in the Supreme Court of Israel that allows each justice to compose three‐justice panels, collecting an original database of decisions in this procedure. The data reveal strong bias in justices’ panel composition. A Gini coefficient measuring the extent of inequality in each justice's panel composition, which runs between 0 (total equality) and 1 (total inequality), is 0.82 on average, which contradicts the random composition theory. The high variance in the choice of panel members contradicts the professional composition theory. The data support the idea that justices compose panels strategically, and accordingly the data uncover justices’ revealed preference for panel members. We use the data to depict the relationships within the Supreme Court of Israel, and identify three groups of justices. Lastly, we show that justices who were selected by the current Chief Justice under the above procedure, before she became Chief Justice, are more likely to sit on panel with her in ordinary hearings after she became Chief Justice. Since the Chief Justice has the legal authority to compose ordinary panels, this is also consistent with strategic panel composition.
O Supremo Tribunal Federal (STF) é uma das instituições mais importantes do país, e tem recebido cada vez mais atenção da sociedade brasileira. Ano a ano, a corte é acionada para julgar dezenas de milhares de casos, muitos deles com impacto direto na vida econômica, política e social do país. Investigar quais são os fatores que influenciam o comportamento dos ministros do STF é fundamental para compreendermos o processo de tomada de decisões do Tribunal. Apesar de existirem estudos importantes sobre o resultado das ações de controle de constitucionalidade, ainda existe espaço para estudar o comportamento individual dos ministros. O objetivo do presente trabalho é analisar o comportamento individual dos ministros do STF, utilizando-se as votações dos ministros frente às ações de controle de constitucionalidade que foram decididas de forma colegiada. A proposta de trabalho é verificar como os ministros se agrupam e quais fatores influenciam sua divisão. Para isso, propomos a utilização do método de estimação de pontos ideias. Nossa hipótese primária de trabalho é que existe uma influência da indicação presidencial no comportamento dos ministros do Supremo Tribunal Federal. Dessa forma, haveria uma divisão da Corte entre os ministros nomeados por diferentes partidos. A hipótese secundária seria que as trajetórias profissionais dos ministros também influenciam em sua forma de decidir, sendo possível, portanto, identificar divisões entre os ministros a partir desta variável. Os resultados do presente trabalho indicam que variáveis associadas aos modelos atitudinais do comportamento judicial, partidos dos presidentes que indicaram os ministros, ideologia e filosofia judicial dos ministros, são variáveis mais relevantes para explicar as agrupamentos e dissensos no STF do que as que dizem respeito as trajetórias profissionais dos juízes.
De uso difundido em estudos sobre comportamento legislativo, técnicas de estimação de pontos ideais passaram a ser utilizadas, nos últimos anos, também para a análise do STF (Supremo Tribunal Federal). Essa nova agenda de pesquisa tem sido útil para identificar padrões de divergência no interior do Tribunal, deslocando, para o comportamento individual, o foco que o campo, por muito tempo, deu ao comportamento coletivo dos ministros. Porém, não existe nenhuma discussão sobre as diversas técnicas disponíveis para essa tarefa e sobre as diferenças que elas podem produzir quando aplicadas aos dados disponíveis sobre o STF. Em decorrência disso, já é possível observar uma considerável variedade de abordagens em uma agenda de pesquisa ainda incipiente. A partir da comparação dos resultados obtidos pela aplicação, a um mesmo conjunto de dados, de quatro entre as principais técnicas de estimação de pontos ideais (Optimal Classification, W-NOMINATE, α-NOMINATE, e PCA), este trabalho busca esclarecer as diferenças entre elas à luz das especificidades dos dados disponíveis sobre as votações do STF, especialmente no julgamento de ações de inconstitucionalidade.
Hybrid constitutional courts are associated with deeply divided and post-conflict contexts where the impartiality of the domestic judiciary is suspect. Such courts enlist international (i.e. foreign) judges to provide an ostensibly neutral counterbalance to the presumed political biases of local judges. This mixed-methods case study of the Constitutional Court of Bosnia-Herzegovina questions the value of these hybrid courts. The results of multidimensional scaling show that, contrary to what might be expected, Bosnia’s foreign judges are not reliable median “swing” judges in divided decisions. Furthermore, qualitative analysis suggests that the foreign judges have been pivotal in several strategic “mistakes” that have probably harmed the Court’s tenuous authority. It is also suggested that the presence of international judges on constitutional courts may actually discourage the kind of strategic behavior that is needed to build and sustain judicial power, particularly in deeply divided and post-conflict contexts.
ResearchGate has not been able to resolve any references for this publication.